2006-08-21T00:00:22 <dreimark> I think I shuld get the version which is now actual
2006-08-21T00:00:47 <dreimark> but if I sync it get's alway my version which is only the first line
2006-08-21T00:00:47 <dreimark> I sync now
2006-08-21T00:01:23 <xorAxAx> Kepplar: still alive?
2006-08-21T00:01:28 <Kepplar> yep
2006-08-21T00:01:56 <xorAxAx> nice to know
2006-08-21T00:02:10 <xorAxAx> dreimark: did you do any local changes?
2006-08-21T00:02:16 <dreimark> no
2006-08-21T00:02:45 <xorAxAx> direction is both?
2006-08-21T00:02:50 <dreimark> yeah
2006-08-21T00:04:25 <xorAxAx> how does the revision no. 5 of that page look like in your local wiki?
2006-08-21T00:04:39 <xorAxAx> hmm, wait
2006-08-21T00:04:45 <xorAxAx> there is no 5 :)
2006-08-21T00:05:39 <xorAxAx> hey, this is a nice test case
2006-08-21T00:05:43 <xorAxAx> i mentally reproduced it :)
2006-08-21T00:05:58 <dreimark> xorAxAx:I am trying now reverting and a change and then syncing this takes time now
2006-08-21T00:05:58 <xorAxAx> you reverted the page behind the first sync and it mixed up the tags
2006-08-21T00:06:14 <xorAxAx> i make a note and will look at it tomorrow
2006-08-21T00:06:55 <xorAxAx> dreimark: how does rev 1 in your local wiki look like?=
2006-08-21T00:07:41 <dreimark> xorAxAx:it's only the first sentence Ask your questions about ...
2006-08-21T00:08:05 <xorAxAx> ok
2006-08-21T00:08:29 <dreimark> Now with the change of the reverted page I've only the first sentence too with the changed words
2006-08-21T00:09:01 <xorAxAx> yeah, something is mixed up
2006-08-21T00:09:19 <xorAxAx> so the first sync trimmed down te page?
2006-08-21T00:09:26 <xorAxAx> or how did you get this one liner page?
2006-08-21T00:09:32 <dreimark> the log of the Sync action page has the complete text
2006-08-21T00:09:40 <xorAxAx> hehe
2006-08-21T00:09:44 <birkenfeld> good night everyone (actually, good coding I guess :->)
2006-08-21T00:09:51 <xorAxAx> gn birkenfeld
2006-08-21T00:09:58 <dreimark> gn birkenfeld
2006-08-21T00:10:18 <Kepplar> I'm getting a 500 error from stroage.www.de
2006-08-21T00:10:36 <dreimark> I got the first line always only if I sync this page
2006-08-21T00:11:00 <birkenfeld> Kepplar: syntax error
2006-08-21T00:11:02 <xorAxAx> dreimark: funny page
2006-08-21T00:11:07 <birkenfeld> a suite may not consist of comments only
2006-08-21T00:11:19 <xorAxAx> birkenfeld: lets play cards!
2006-08-21T00:11:29 <birkenfeld> so if you comment out the only code line in a suite, bang
2006-08-21T00:11:56 <birkenfeld> xorAxAx: Schafkopf anyone?
2006-08-21T00:12:09 <xorAxAx> birkenfeld: which suites do you have there? :)
2006-08-21T00:12:32 * birkenfeld is looking up Schell in English...
2006-08-21T00:12:57 <dreimark> xorAxAx:It was the largest page MM has
2006-08-21T00:13:13 <xorAxAx> dreimark: thats quite funny ...
2006-08-21T00:13:30 <xorAxAx> dreimark: you created it remotely, started sync and it trimmed it down, right?
2006-08-21T00:13:43 <dreimark> yeah
2006-08-21T00:14:01 <xorAxAx> ok
2006-08-21T00:14:36 <dreimark> To try again I've removed the entire dir from pages
2006-08-21T00:15:09 <xorAxAx> yeah, it might have had incorrect tags
2006-08-21T00:15:15 <xorAxAx> and, did that help?
2006-08-21T00:15:23 <dreimark> no
2006-08-21T00:15:33 <dreimark> I do see the first line
2006-08-21T00:16:59 <dreimark> try again now have not reverted before
2006-08-21T00:20:23 <xorAxAx> dreimark: please either prune your whole data dir or change the page name
2006-08-21T00:20:35 <xorAxAx> you are confusing the sync system
2006-08-21T00:20:43 <xorAxAx> i.e. example4
2006-08-21T00:23:46 <Kepplar> ThomasWaldmann: what do i Do?
2006-08-21T00:23:54 <Kepplar> ThomasWaldmann: its causing too many errors.. this is JUST for page integration
2006-08-21T00:24:02 <Kepplar> ThomasWaldmann: please see storage.www.de to get my point
2006-08-21T00:24:07 <Kepplar> ThomasWaldmann: ive been saying this for weeks
2006-08-21T00:24:21 <dreimark> xorAxAx:all examples removed reverted and sync started
2006-08-21T00:24:34 <Kepplar> ThomasWaldmann: i18n has NOTHING to do with storage
2006-08-21T00:24:56 <xorAxAx> Kepplar: there are system wide caches in cfg.cache_dir
2006-08-21T00:25:04 <xorAxAx> they are not related to page storage
2006-08-21T00:25:18 <Kepplar> i have *no* idea how this cache works
2006-08-21T00:25:25 <Kepplar> but it keeps calling Page
2006-08-21T00:25:42 <xorAxAx> dreimark: using a new pagename please
2006-08-21T00:25:45 <xorAxAx> example5 or so
2006-08-21T00:25:48 <xorAxAx> 3 and 4 are used
2006-08-21T00:25:52 <Kepplar> for methods that don't exist anymore and cannot exist anymore
2006-08-21T00:26:19 <xorAxAx> Kepplar: please try to understand te code
2006-08-21T00:26:28 <xorAxAx> it only calls page if the arena is page related
2006-08-21T00:26:39 <Kepplar> yes i know that
2006-08-21T00:26:52 <Kepplar> ive gone through and commented out anything related
2006-08-21T00:27:20 <xorAxAx> obviously not the right things :)
2006-08-21T00:27:33 <Kepplar> well no, now the problem is something that called CACHE
2006-08-21T00:27:35 <xorAxAx> now lets calculate how many permutations of commented out code lines exist
2006-08-21T00:27:44 <birkenfeld> xorAxAx: lol
2006-08-21T00:27:57 <Kepplar> i don't have time to read and understand cache
2006-08-21T00:28:05 <Kepplar> its stupidly under-documented
2006-08-21T00:28:19 <birkenfeld> looks like it is not industry standard
2006-08-21T00:28:38 <Kepplar> Rubik file is what documentation should look like :)
2006-08-21T00:28:43 <Kepplar> and thats probably not complete
2006-08-21T00:29:02 <xorAxAx> Kepplar: just remove all your comments but the one in line 40 and it should work
2006-08-21T00:30:09 <Kepplar> nope
2006-08-21T00:30:12 <Kepplar> i did that at the start
2006-08-21T00:30:18 <Kepplar> firstly line 37 is the same thing
2006-08-21T00:30:32 <Kepplar> that cascades to errors on 50-53 area
2006-08-21T00:30:37 <xorAxAx> hmm, then i should suggest another permutation maybe
2006-08-21T00:30:37 <Kepplar> which then takes out two entire methofs
2006-08-21T00:30:44 <birkenfeld> why not remove caching.py altogether?
2006-08-21T00:30:59 <Kepplar> birkenfeld: tbh it probably is now deprecated
2006-08-21T00:31:16 <Kepplar> however i have no idea how to remove it
2006-08-21T00:31:47 <xorAxAx> no, its not deprecated
2006-08-21T00:31:52 <xorAxAx> moin will always need caches
2006-08-21T00:31:59 <Kepplar> i know
2006-08-21T00:32:10 <Kepplar> but all disk access should go via storage
2006-08-21T00:32:15 <xorAxAx> no!
2006-08-21T00:32:24 <xorAxAx> caching doesnt need to to get it running first
2006-08-21T00:32:39 <Kepplar> oh i know that
2006-08-21T00:32:42 <Kepplar> i mean down the line
2006-08-21T00:32:43 <xorAxAx> so just modify the path dispatcer in __init__ to write to cache_dir always
2006-08-21T00:33:41 * Kepplar looks blankly
2006-08-21T00:34:29 <Kepplar> i dont see anything called dispatcher
2006-08-21T00:34:32 <dreimark> xorAxAx:Example5 doesn't work too
2006-08-21T00:34:41 <dreimark> only first sentence
2006-08-21T00:34:46 <xorAxAx> ok, nice
2006-08-21T00:34:47 <birkenfeld> perhaps now is the time for a nice bottle of vodka...
2006-08-21T00:34:52 <xorAxAx> will reproduce it tomorrow
2006-08-21T00:39:39 <dreimark> gn
2006-08-21T00:40:56 <xorAxAx> gn dreimark
2006-08-21T01:18:19 <xorAxAx> Kepplar: have you realised that we are allowed to help you after you have finished SOC?
2006-08-21T01:22:36 <xorAxAx> thats actually nice
2006-08-21T10:35:56 <Kepplar> moin
2006-08-21T10:40:21 <xorAxAx> moin
2006-08-21T10:43:40 <Kepplar> no milk ://
2006-08-21T10:43:50 * Kepplar sighs
2006-08-21T10:43:51 <Kepplar> coffee it is
2006-08-21T11:03:41 <ThomasWaldmann> moin
2006-08-21T11:03:50 <birkenfeld> moin
2006-08-21T11:03:51 <Kepplar> moin
2006-08-21T11:04:34 <ThomasWaldmann> ok, lets do our last SOC monday morning meeting :)
2006-08-21T11:04:50 <xorAxAx> is fpletz here?
2006-08-21T11:04:51 <ThomasWaldmann> i guess we can skip the "next week" part today :)
2006-08-21T11:05:07 <xorAxAx> ThomasWaldmann: why? people might want to continue to work on it
2006-08-21T11:05:11 <Kepplar> yea
2006-08-21T11:05:13 <ThomasWaldmann> maybe he arrives later
2006-08-21T11:05:30 <ThomasWaldmann> i said "SOC", sure can people continue, but it is not SOC then :)
2006-08-21T11:05:41 <Kepplar> AOF
2006-08-21T11:05:46 <ThomasWaldmann> xorAxAx: can you begin today?
2006-08-21T11:05:50 <xorAxAx> yes
2006-08-21T11:05:55 <xorAxAx> LAST: Started designing the solutions for the other sync cases. Store and transmit the normalised name. Implemented preliminary mime type support, only transmission of the mime type and the new storage API is missing. Changed getDiff interface. Detect renamed pages (and choke :). Added infrastructure support for detecting deleted pages. Reworked i18n and logging support. Added GetText2 macro. Getting syncing of pages working that ...
2006-08-21T11:06:02 <xorAxAx> ... are just available on one side. Working synchronisation of deleted pages. Implemented rollback in case of remote problems and exception logging.
2006-08-21T11:06:04 <xorAxAx> NEXT: Add user documentation to MoinMaster. Try to implement a MultiCall reactor that automatically collects and batches all RPC calls in order to reduce load and latency.
2006-08-21T11:06:07 <xorAxAx> BLOCKERS: Hmm, a usable new storage subsystem would be nice.
2006-08-21T11:06:12 <xorAxAx> note that NEXT describes the tasks that i want to finish in the next 5 hours :)
2006-08-21T11:07:03 <ThomasWaldmann> choke?
2006-08-21T11:07:11 <xorAxAx> yes, its not supported
2006-08-21T11:07:16 <xorAxAx> so it gives an error message
2006-08-21T11:07:23 <xorAxAx> (not a traceback :))
2006-08-21T11:07:52 <xorAxAx> for locally renamed pages, it should even continue to sync (but it doesnt rename the page)
2006-08-21T11:08:28 <ThomasWaldmann> so if it got renamed on the server while client was away, you will have to do the same rename on the client manually?
2006-08-21T11:09:15 <xorAxAx> no, you will have to delete the page and start over
2006-08-21T11:09:28 <xorAxAx> because are not supported, by design :)
2006-08-21T11:09:34 <xorAxAx> s/are/renames &/
2006-08-21T11:10:17 <xorAxAx> because the sync system doesnt know anything about the page history of a remote page
2006-08-21T11:10:27 <xorAxAx> so it cannot determine when a page was renamed how
2006-08-21T11:11:11 <ThomasWaldmann> ok
2006-08-21T11:11:15 <xorAxAx> currently it just detects the case when pages were renamed because that gives pagenames in the tags that dont match
2006-08-21T11:11:27 <xorAxAx> i am sure that this limitation can be removed, but it would require some rename tracking
2006-08-21T11:11:42 <xorAxAx> and an RPC action that reads the changelog
2006-08-21T11:11:47 <xorAxAx> umm, global edit-log
2006-08-21T11:12:06 <xorAxAx> or some other tricks (described in a doc string)
2006-08-21T11:13:13 <ThomasWaldmann> btw, I think cleanup and docs are more important than multicall stuff
2006-08-21T11:13:44 <xorAxAx> the code is clean, just docs are missing
2006-08-21T11:13:48 <xorAxAx> and yes, i will tackle them first
2006-08-21T11:14:49 <ThomasWaldmann> CHANGES.* should be in a state parts of it can be copied 1:1 to main CHANGES
2006-08-21T11:15:46 <xorAxAx> ok
2006-08-21T11:15:57 <ThomasWaldmann> GetText2 does what?
2006-08-21T11:16:23 <xorAxAx> it translates the first parameter and replaces %s by the other parameters
2006-08-21T11:16:33 <xorAxAx> so you can use it for messages that contain data
2006-08-21T11:16:47 <ThomasWaldmann> ah, that stuff
2006-08-21T11:17:05 <ThomasWaldmann> ffesti: are you here? (just checking :)
2006-08-21T11:18:05 <ThomasWaldmann> xorAxAx: maybe try to chase up your mentor :)
2006-08-21T11:18:28 <xorAxAx> heh, i remember my chasing for the mid-term evals
2006-08-21T11:18:59 <xorAxAx> he might not yet be in the office
2006-08-21T11:19:32 <xorAxAx> or he has some meeting, who knows
2006-08-21T11:21:07 <ThomasWaldmann> remove the Developer notes in CHANGES, if they stay as they are.
2006-08-21T11:21:26 <xorAxAx> sure
2006-08-21T11:22:16 <ThomasWaldmann> fpletz: are you here now?
2006-08-21T11:23:14 <ThomasWaldmann> some people really need some preemptive scheduling
2006-08-21T11:23:42 <ThomasWaldmann> xorAxAx: any last words? <g>
2006-08-21T11:23:46 <xorAxAx> no
2006-08-21T11:23:57 <xorAxAx> yummy peanut creme
2006-08-21T11:24:00 <ThomasWaldmann> ok, kepplar, your time :)
2006-08-21T11:24:11 <Kepplar> Hi
2006-08-21T11:24:14 <Kepplar> Hmm
2006-08-21T11:24:18 <Kepplar> LAST - Integration
2006-08-21T11:24:21 <Kepplar> NEXT - Integration
2006-08-21T11:24:35 <ThomasWaldmann> heh
2006-08-21T11:24:44 <ThomasWaldmann> does anything practically work yet?
2006-08-21T11:24:46 <Kepplar> BLOCKERS - You already tknow
2006-08-21T11:24:53 <Kepplar> I think that was the mistake
2006-08-21T11:25:11 <Kepplar> I think Trivial end 2 end storage tests should have been completed
2006-08-21T11:25:23 <Kepplar> as now I'm jumping around trying to work out where the bug is
2006-08-21T11:25:43 <Kepplar> main problems is I'm not sure retriving data is working
2006-08-21T11:25:48 <Kepplar> and acl
2006-08-21T11:25:56 <Kepplar> users do save to disk
2006-08-21T11:26:06 <Kepplar> but unable to retrive the correct uid
2006-08-21T11:26:10 <ThomasWaldmann> acl shouldnt be a problem as you have disabled it
2006-08-21T11:26:22 <Kepplar> thats fair enough is i knew HOW they worked
2006-08-21T11:26:58 <Kepplar> Its like saying disabled the <insert random technical sounding word> from a car engine and it should work when you've never touched a car engine
2006-08-21T11:26:58 <xorAxAx> Kepplar: getACL has to return an AccessControlList object
2006-08-21T11:27:02 <xorAxAx> nothing more than that
2006-08-21T11:27:17 <xorAxAx> and it needs to call the ACL code with the ACL line for that
2006-08-21T11:27:19 <xorAxAx> rather trivial
2006-08-21T11:27:25 <ThomasWaldmann> the uid is in the cookie, you just have to get the respective file from disk (for flatfile)
2006-08-21T11:27:36 <Kepplar> xorAxAx: but it IS returning none thats causing the error
2006-08-21T11:27:53 <Kepplar> the user uid's are being fetched from disk upon lookup
2006-08-21T11:28:02 <Kepplar> its the parsing of it to match it with the uid to get the name which is a problem
2006-08-21T11:28:34 <ThomasWaldmann> we have a name2id mapping cached
2006-08-21T11:28:37 <Kepplar> I simply don't get why getPagePath has left exposed to everything
2006-08-21T11:28:54 <Kepplar> i saw that mentioned - i have no idea how it works
2006-08-21T11:28:57 <xorAxAx> Kepplar: thats neither related to ACL nor to user
2006-08-21T11:28:58 <Kepplar> (couldnt find any documentation)
2006-08-21T11:29:15 <ThomasWaldmann> getpagepath for user context returns only data_dir
2006-08-21T11:29:18 <Kepplar> xorAxAx: of course it is, if you can't get the userlist you can't lookup a user by name
2006-08-21T11:29:27 <Kepplar> xorAxAx: which is what its trying to do
2006-08-21T11:29:30 <xorAxAx> Kepplar: but the user doesnt use getPagePath
2006-08-21T11:29:34 <xorAxAx> Kepplar: its not related
2006-08-21T11:29:41 <Kepplar> xorAxAx: thats ANOTHER error
2006-08-21T11:29:54 <xorAxAx> no, thats expected because users are not related to pages
2006-08-21T11:30:10 <Kepplar> ....
2006-08-21T11:30:13 <Kepplar> what are you talking about?
2006-08-21T11:30:24 <xorAxAx> about you claiming that getpagepath is everywhere
2006-08-21T11:30:30 <Kepplar> yes
2006-08-21T11:30:32 <ThomasWaldmann> if getpagepath is called, it is just called on the virtual rootpage -and this just returns the data_dir (as mentioned about 5x already)
2006-08-21T11:30:46 <Kepplar> ah
2006-08-21T11:30:52 <Kepplar> except its not allowed to know about that
2006-08-21T11:30:56 <Kepplar> getpagepath has been removed weeks ago
2006-08-21T11:31:13 <ThomasWaldmann> expected result: .../data/user, EVER
2006-08-21T11:31:31 <Kepplar> thats not allowed to know about it
2006-08-21T11:31:54 <ThomasWaldmann> (and i told you to implement some dummy rootpage object just doing that stuff)
2006-08-21T11:31:58 <xorAxAx> Kepplar: you are confusing levels
2006-08-21T11:32:04 <Kepplar> i have done so
2006-08-21T11:32:05 <xorAxAx> Kepplar: your backend has to know about it
2006-08-21T11:32:11 <Kepplar> yes
2006-08-21T11:32:12 <Kepplar> it does
2006-08-21T11:32:22 <xorAxAx> the other code hasnt to
2006-08-21T11:32:27 <Kepplar> i know
2006-08-21T11:32:32 <xorAxAx> so just rip out the code and call your stuff
2006-08-21T11:32:33 <Kepplar> thats what the intended result is
2006-08-21T11:32:43 <Kepplar> xorAxAx: thats what im doing...
2006-08-21T11:32:57 <xorAxAx> what you have been doing for 3 weeks, yep
2006-08-21T11:33:10 <Kepplar> because its so extensive and undocumented
2006-08-21T11:33:35 <Kepplar> and now flatfile is a giant hack
2006-08-21T11:33:59 <xorAxAx> hack == design is broken :)
2006-08-21T11:34:05 <Kepplar> correct
2006-08-21T11:34:08 <Kepplar> legacy is crap
2006-08-21T11:34:17 <xorAxAx> i.e. it doesnt fit the existing system
2006-08-21T11:34:25 <Kepplar> imho the current system is wrong
2006-08-21T11:34:26 <xorAxAx> i.e. it wasnt considered to make it fit
2006-08-21T11:34:32 <ThomasWaldmann> maybe if you replace everything not working by senseful dummy function, you could get at least a working wiki (no acls, no logged in user)
2006-08-21T11:34:44 <Kepplar> already tried that
2006-08-21T11:34:46 <Kepplar> didnt worka
2006-08-21T11:34:50 <Kepplar> -a
2006-08-21T11:34:57 <ThomasWaldmann> that would be better than just showing tracebacks
2006-08-21T11:35:10 <Kepplar> all nice in theory
2006-08-21T11:35:32 <Kepplar> all references to getPathPath must be migrated there is no way around it
2006-08-21T11:35:40 <Kepplar> I cannot supply that information to put it simply
2006-08-21T11:35:42 <Kepplar> impossible
2006-08-21T11:36:04 <Kepplar> blank or none will cause it to break
2006-08-21T11:36:06 <Kepplar> i tried that
2006-08-21T11:36:58 <ffesti> moin
2006-08-21T11:37:02 <ThomasWaldmann> there aren't many
2006-08-21T11:37:19 <Kepplar> a few dozens
2006-08-21T11:37:27 <Kepplar> each one requriing extensive code changes to accomdate it
2006-08-21T11:37:39 <Kepplar> plus the fact I'm not familar with how any of these subsystems work of fit in
2006-08-21T11:38:32 <ThomasWaldmann> edit-log: it is either referencing the global edit-log in data/ or you don't need it as you store single log entries in your meta data
2006-08-21T11:38:56 <Kepplar> yes, but knowing what it does and HOW it works are two different thing
2006-08-21T11:39:12 <ThomasWaldmann> event-log: we only have one, it is data/event-log
2006-08-21T11:39:25 <Kepplar> event log ive left out
2006-08-21T11:39:44 <Kepplar> that needs to be put into storage and i haven't even had the chance to get around to implimenting it in storage
2006-08-21T11:39:56 <xorAxAx> Kepplar: it all boils down to "you have to understand the callers and, furthermore, the design has to reflect the callees usages"
2006-08-21T11:40:12 <Kepplar> xorAxAx: but its impossible
2006-08-21T11:40:16 <xorAxAx> no
2006-08-21T11:40:24 <Kepplar> yes
2006-08-21T11:40:31 <xorAxAx> why?
2006-08-21T11:40:51 <xorAxAx> its perfectly possible in all cases you talked about this morning
2006-08-21T11:40:54 <ThomasWaldmann> caching: just make up some path "per page" on the disk (use cache_dir as basedir)
2006-08-21T11:41:02 <Kepplar> an abstract storage engine which can have a range of storage implimentations to provide a storage neutral api CANNOT SUPPLY a filesystem uri
2006-08-21T11:41:05 <Kepplar> path
2006-08-21T11:41:26 <xorAxAx> Kepplar: i am not talking about implementing that method but about rewriting the caller of course
2006-08-21T11:41:41 <Kepplar> what callers?
2006-08-21T11:41:47 <xorAxAx> the callers that call getpagepath
2006-08-21T11:41:51 <xorAxAx> (for example)
2006-08-21T11:41:51 <Kepplar> ....
2006-08-21T11:41:54 <Kepplar> thats what ive been doing
2006-08-21T11:41:54 <ThomasWaldmann> PageEditor: ignore rename stuff, comment out underlay related stuff
2006-08-21T11:42:02 <xorAxAx> Kepplar: so it is not impossible
2006-08-21T11:42:23 <Kepplar> xorAxAx: thats precisely what ive been saying
2006-08-21T11:42:39 <xorAxAx> 11:40:12 < Kepplar> xorAxAx: but its impossible
2006-08-21T11:42:42 <Kepplar> and have i mentioned the lack of documentation doesnt help at all?
2006-08-21T11:42:56 <xorAxAx> Kepplar: you may ask your mentor, me or other peeps here :)
2006-08-21T11:42:57 <Kepplar> xorAxAx: you were talking about storage design
2006-08-21T11:43:02 <Kepplar> that has nothing to do with storage design
2006-08-21T11:43:03 <xorAxAx> (and even try to understand the code)
2006-08-21T11:43:08 <ThomasWaldmann> Page loadCache: dont check for cacheneedsupdate
2006-08-21T11:43:19 <xorAxAx> Kepplar: oh, it has. you need some method that the callers can call in your system
2006-08-21T11:43:28 <ThomasWaldmann> i guess this is about all you need to get around getpagepath
2006-08-21T11:43:32 <xorAxAx> Kepplar: e.g. "LoggingItem.attach_record(...)"
2006-08-21T11:43:50 <Kepplar> im not talking about that
2006-08-21T11:43:54 <Kepplar> im talking about getpagepath
2006-08-21T11:44:12 <xorAxAx> or sem.create_new_item("Cache/.../my_cache_item")
2006-08-21T11:44:18 <xorAxAx> etc. pp. :)
2006-08-21T11:44:23 <Kepplar> ....
2006-08-21T11:44:28 <Kepplar> i said caching wasnt implimented yet
2006-08-21T11:45:00 <ThomasWaldmann> you don't need to do caching via your stuff
2006-08-21T11:45:01 <Kepplar> and quite frankly caching should be transparent
2006-08-21T11:45:13 <Kepplar> xorAxAx brought it up
2006-08-21T11:45:48 <xorAxAx> Kepplar: my first advise was not to use storage for that yesterday
2006-08-21T11:46:08 <Kepplar> .........
2006-08-21T11:46:08 <xorAxAx> Kepplar: i said that you should rewrite the caching__init__ to build sensible paths for page objects
2006-08-21T11:46:15 <xorAxAx> that would have solved this problem
2006-08-21T11:46:17 <Kepplar> correct
2006-08-21T11:46:21 <Kepplar> But
2006-08-21T11:46:25 <Kepplar> if you've been listening at all
2006-08-21T11:46:38 <Kepplar> you might have realised i said about a billion times I dont have the time to understand every subsystem
2006-08-21T11:46:51 <ThomasWaldmann> cache_dir + quoteWikinameFS(name)
2006-08-21T11:47:02 <xorAxAx> but that is necessary, Kepplar
2006-08-21T11:47:10 <ThomasWaldmann> cache_dir + '/pages/' + quoteWikinameFS(name)
2006-08-21T11:47:20 <Kepplar> ok
2006-08-21T11:47:23 <Kepplar> I can't do that
2006-08-21T11:47:24 <xorAxAx> and it shouldnt need more than 2 weeks to understand moin well enough while playing with its source code
2006-08-21T11:47:39 <Kepplar> im not a reverse engineer
2006-08-21T11:47:51 <Kepplar> (Thank god)
2006-08-21T11:48:01 <ThomasWaldmann> fpletz: are you there?
2006-08-21T11:48:05 <xorAxAx> heh, but you are expected to be one :)
2006-08-21T11:48:19 <xorAxAx> Kepplar: on one project, we dont have any comments or dev docs at all
2006-08-21T11:48:27 <xorAxAx> Kepplar: its written in java and the code is horrible
2006-08-21T11:48:34 <Kepplar> i'd sack whoever wrote it
2006-08-21T11:48:41 <xorAxAx> nevertheless, even people that studied economics get into the code
2006-08-21T11:48:50 <Kepplar> thats fine
2006-08-21T11:48:52 <Kepplar> if
2006-08-21T11:49:06 <xorAxAx> they use e.g. tracing through it
2006-08-21T11:49:16 <richardb> Hmm, according to lastfm, fpletz was listening to music 6 hours ago... I'm guessing he's asleep.
2006-08-21T11:49:16 <Kepplar> you have undocumented legacy code and the ethos was all new code is documented and the old code is slowly being upgraded
2006-08-21T11:49:18 <xorAxAx> thats barely possible if you do not have a local installation
2006-08-21T11:49:24 <xorAxAx> richardb: heh
2006-08-21T11:49:45 <ffesti> may be it is more clever to discuss how to get stuff working...
2006-08-21T11:50:05 <xorAxAx> ffesti: i guess you didnt follow the discussions in the last days :)
2006-08-21T11:50:09 <ThomasWaldmann> Kepplar: i suggest you systematically fix one problem after the other now (based on what backtraces you get when doing practical tests)
2006-08-21T11:50:24 * ffesti was afk over the WE
2006-08-21T11:50:32 <Kepplar> xorAxAx: on my uml docuuments i have "here be dragons" for non-storage moin code beyond page/user etc
2006-08-21T11:50:40 <Kepplar> xorAxAx: code should have clearly defined apis
2006-08-21T11:50:41 <ThomasWaldmann> if a fix isn't obvious to you and you can't find a hint some lines above here, ask.
2006-08-21T11:50:59 <xorAxAx> Kepplar: thats bad - integrate and dont suppress
2006-08-21T11:50:59 <Kepplar> k
2006-08-21T11:51:19 <Kepplar> xorAxAx: you not done component programming then O-o
2006-08-21T11:51:35 <Kepplar> ?
2006-08-21T11:51:43 <xorAxAx> Kepplar: modelling a component with dragons is not very goal-oriented :)
2006-08-21T11:51:51 <Kepplar> well
2006-08-21T11:51:53 <Kepplar> thats non-storage
2006-08-21T11:52:09 <xorAxAx> if something is storing data, it is considered to be related to storage
2006-08-21T11:52:10 <Kepplar> i expected to be honest that anything behind page didnt etc call storage directly
2006-08-21T11:52:23 <ThomasWaldmann> (and please accept that if you job is doing a clean api somewhere, then you have to expect that it is not clean yet, or it wouldn't be your job)
2006-08-21T11:52:55 <Kepplar> oh sure i know that
2006-08-21T11:53:04 <Kepplar> but i didn't realise how extensive it was
2006-08-21T11:53:17 <Kepplar> i thought the only things I'd be touching was user, Page and Page Editor
2006-08-21T11:53:43 <ThomasWaldmann> i told you about 1-2 months ago that you aren't proceeding fast enough, but you told that you are "in time" :)
2006-08-21T11:53:54 <Kepplar> to the plan
2006-08-21T11:54:31 <ThomasWaldmann> maybe just better follow what people more experienced with the stuff tell you.
2006-08-21T11:54:40 <Kepplar> i was going my fastest
2006-08-21T11:55:36 <Kepplar> code quality imho is most important
2006-08-21T11:55:38 <ThomasWaldmann> fpletz: ok, it looks like you missed this meeting. just another scheduling problem... - please speak up when you return.
2006-08-21T11:56:41 <ThomasWaldmann> richardb: are you there? student projects deadline is today at 15:00 UTC.
2006-08-21T11:56:53 <richardb> Yes, I'm here.
2006-08-21T11:57:17 <ThomasWaldmann> did you see latest stuff from fpletz?
2006-08-21T11:57:41 <richardb> I've been looking at his latest commits, yes.
2006-08-21T11:58:23 <richardb> Haven't tried it out myself recently, though. Am doing so at the moment.
2006-08-21T11:58:44 <ThomasWaldmann> http://test.wikiwikiweb.de/AdvancedSearch (server is a bit slow)
2006-08-21T11:59:12 <richardb> Ah, yes, I've tried that.
2006-08-21T11:59:20 <richardb> I meant, I'm trying a local install.
2006-08-21T11:59:35 <ThomasWaldmann> language selection seems to not work
2006-08-21T12:00:22 <ThomasWaldmann> file type maybe works, but UI is maybe not what a normal user can cope with
2006-08-21T12:00:38 <ThomasWaldmann> didnt try last-modified search
2006-08-21T12:00:48 <richardb> The UI looks okay to me...
2006-08-21T12:01:03 <ThomasWaldmann> (usage is unclear, XXX?)
2006-08-21T12:01:13 <richardb> Were talking about the advanced search UI, I assume?
2006-08-21T12:01:34 <richardb> Ah - the last modified bit, yes.
2006-08-21T12:01:51 <richardb> That needs to explain what kind of input can be entered in the form input.
2006-08-21T12:02:31 <richardb> Might be better if some common file types were displayed at the start of the filetype list.
2006-08-21T12:02:44 <richardb> Or if only the filetypes actually present in the database were listed.
2006-08-21T12:03:00 <birkenfeld> advise is needed for the terms boxes too
2006-08-21T12:03:09 <birkenfeld> entering ".*" results in a traceback
2006-08-21T12:03:28 <richardb> Oops.
2006-08-21T12:04:43 <birkenfeld> doing it with the first box gives a different error compared to the second one
2006-08-21T12:04:58 <birkenfeld> submitting the form without any text entered gives yet another traceback
2006-08-21T12:05:13 <xorAxAx> %-)
2006-08-21T12:05:30 <ThomasWaldmann> somehow the code on xapian.www.de doesnt work like expected (language sorting, see last commit)
2006-08-21T12:05:40 <birkenfeld> I don't want to be a joykiller though
2006-08-21T12:05:56 <richardb> birkenfeld: *sigh* Those should all be easy to fix... given time. (!)
2006-08-21T12:06:00 <xorAxAx> birkenfeld: tests are important :)
2006-08-21T12:06:46 <birkenfeld> I wonder who creates a HTML form and doesn't check out some corner cases
2006-08-21T12:07:50 <xorAxAx> users are well-meaning :)
2006-08-21T12:08:07 <xorAxAx> people storing things in moin are well-meaning
2006-08-21T12:08:42 <ThomasWaldmann> except that guy yesterday on test.www.de
2006-08-21T12:10:34 <xorAxAx> (of course they are not, but people often assume that, which is wrong)
2006-08-21T12:11:00 <richardb> Hmm - do we have to send stuff to google by the deadline today?
2006-08-21T12:11:06 <richardb> What exactly does the deadline mean?
2006-08-21T12:11:09 <xorAxAx> richardb: they want a pointer
2006-08-21T12:11:16 <xorAxAx> and there are rumors that it was moved
2006-08-21T12:11:18 <xorAxAx> by one day
2006-08-21T12:11:48 <richardb> Well, don't rely on that unless they become more than rumours...
2006-08-21T12:12:05 <xorAxAx> LH wrote that the evals are moved
2006-08-21T12:12:28 <xorAxAx> but that doesnt mean that the mentors cannot enforce the deadline on their own
2006-08-21T12:12:47 <birkenfeld> <insert evil laughter> ?
2006-08-21T12:13:19 <xorAxAx> hehe
2006-08-21T12:14:01 <richardb> Ah, I see: "Part of the final surveys for mentors will be a request for a pointer to your student's source code, which will suffice for the final sourcecode upload."
2006-08-21T12:16:13 <ThomasWaldmann> btw, did the $2500 already reach our soc students?
2006-08-21T12:16:30 <xorAxAx> yes
2006-08-21T12:16:43 <xorAxAx> i got my check 5 days ago
2006-08-21T12:16:54 <xorAxAx> i just need to cash it
2006-08-21T12:17:19 * ThomasWaldmann would cash it really soon :)
2006-08-21T12:17:50 <Kepplar> ok
2006-08-21T12:17:51 <Kepplar> so
2006-08-21T12:17:59 <Kepplar> i've removed the page update thing
2006-08-21T12:18:08 <Kepplar> so now onto cachine
2006-08-21T12:18:11 <Kepplar> how exactly does it work?
2006-08-21T12:18:25 <ThomasWaldmann> so no matter what happens soon, you had a nice summer job for $2500+ :)
2006-08-21T12:18:27 <xorAxAx> ThomasWaldmann: its an american cheque, they dont expire if you mean that :)
2006-08-21T12:19:50 <ThomasWaldmann> but could be revoked? google could go bankrupt? they take 2 weeks anyway to cash? ...
2006-08-21T12:20:13 <Kepplar> So theres a central cache file?
2006-08-21T12:20:18 <Kepplar> or cachedir?
2006-08-21T12:20:29 <ThomasWaldmann> request.cfg.cache_dir
2006-08-21T12:20:30 <Kepplar> if so why is there two ways to locate it?
2006-08-21T12:20:42 <Kepplar> arena.getPagePath('cache', check_create=1)
2006-08-21T12:20:45 <Kepplar> why that then?
2006-08-21T12:20:56 <xorAxAx> ThomasWaldmann: no, its directly cached i.V.
2006-08-21T12:20:56 <ThomasWaldmann> there is global cache (farm), per wiki cache, per page cache
2006-08-21T12:21:01 <xorAxAx> valuta +3
2006-08-21T12:21:10 <xorAxAx> because its citybank germany
2006-08-21T12:21:12 <Kepplar> also
2006-08-21T12:21:12 <xorAxAx> (for me)
2006-08-21T12:21:16 <Kepplar> what does cache actually mean?
2006-08-21T12:21:18 <Kepplar> caching what?
2006-08-21T12:21:23 <Kepplar> if its to disk it isnt going to be faster
2006-08-21T12:21:34 <ThomasWaldmann> see MoinMoin:ReducingIO for an oversight
2006-08-21T12:21:55 <ThomasWaldmann> (and sure it's faster :)
2006-08-21T12:24:10 <ThomasWaldmann> and it got tuned for posix by me recently, btw
2006-08-21T12:24:18 <Kepplar> that page doesnt actually make much sense
2006-08-21T12:25:07 <xorAxAx> then ask
2006-08-21T12:25:12 <Kepplar> ok
2006-08-21T12:25:17 <Kepplar> what does caching cache?
2006-08-21T12:25:45 * ThomasWaldmann bbl
2006-08-21T12:25:52 <xorAxAx> caching == retain source data in a processed form that is more quickly to load than the processing needed
2006-08-21T12:26:10 <Kepplar> so formatter output?
2006-08-21T12:26:21 <xorAxAx> moin caches quite a few things
2006-08-21T12:26:32 <xorAxAx> rendered translations, pages in bytecode, etc.
2006-08-21T12:26:39 <Kepplar> bytecode?
2006-08-21T12:26:44 <xorAxAx> yes, bytecode
2006-08-21T12:26:53 <xorAxAx> python has a VM that interprets bytecode
2006-08-21T12:26:59 <Kepplar> python tbyecode?
2006-08-21T12:27:02 <xorAxAx> yep
2006-08-21T12:27:04 <Kepplar> im not stupid :P
2006-08-21T12:27:07 <Kepplar> byte&
2006-08-21T12:27:08 <Kepplar> *
2006-08-21T12:27:09 <Kepplar> ggerfsda
2006-08-21T12:27:17 <Kepplar> i type badly when im stressed :]
2006-08-21T12:27:24 <xorAxAx> so it can pass the bytecode directly to the vm
2006-08-21T12:27:32 <Kepplar> so its pickling data?
2006-08-21T12:27:34 <xorAxAx> which is very fast
2006-08-21T12:27:39 <xorAxAx> no, even faster than pickling
2006-08-21T12:27:44 <xorAxAx> its using marshal
2006-08-21T12:27:53 <Kepplar> why don't we do that for everything?
2006-08-21T12:27:54 <xorAxAx> pickling doesnt give you bytecode
2006-08-21T12:28:03 <xorAxAx> what else comes to your mind?
2006-08-21T12:28:06 <xorAxAx> its used for every page
2006-08-21T12:28:07 <Kepplar> storage?
2006-08-21T12:28:14 <Kepplar> instead of saving text
2006-08-21T12:28:19 <xorAxAx> ?
2006-08-21T12:28:33 <xorAxAx> what should do the bytecode then?
2006-08-21T12:28:48 <Kepplar> I'm saying what shouldn't everything be stored in bytecode
2006-08-21T12:28:54 <Kepplar> why*
2006-08-21T12:29:04 <xorAxAx> because it doesnt make sense
2006-08-21T12:29:20 <xorAxAx> storing in bytecode means that you can only execute the cache item
2006-08-21T12:29:26 <xorAxAx> but you cannot regain text from it
2006-08-21T12:29:52 <Kepplar> its bytecode, its an object inside there perhaps, an object that can be accessed
2006-08-21T12:30:11 <xorAxAx> the bytecode could build a string, yeah
2006-08-21T12:30:19 <xorAxAx> but there is no advantage doing so
2006-08-21T12:30:27 <xorAxAx> because it would be more expensive than reading the string directly
2006-08-21T12:30:31 <Kepplar> i thought you said it was quicker?
2006-08-21T12:30:36 * xorAxAx sighs
2006-08-21T12:30:42 <xorAxAx> i never said that
2006-08-21T12:30:55 <xorAxAx> its quicker to execute bytecode than to render a page
2006-08-21T12:31:10 <Kepplar> 11:27 < xorAxAx> so it can pass the bytecode directly to the vm
2006-08-21T12:31:10 <Kepplar> 11:27 < Kepplar> so its pickling data?
2006-08-21T12:31:12 <xorAxAx> which involves parsing and formatter ops
2006-08-21T12:32:08 <xorAxAx> esp. parsing is expensiv
2006-08-21T12:32:09 <xorAxAx> e
2006-08-21T12:32:09 <Kepplar> shmm
2006-08-21T12:32:22 <Kepplar> parsing what?
2006-08-21T12:32:25 <xorAxAx> the page
2006-08-21T12:32:41 <Kepplar> ah
2006-08-21T12:32:45 <xorAxAx> thats not storage related but nevertheless interesting
2006-08-21T12:33:42 <Kepplar> tw didnt give me a solution for per page cache
2006-08-21T12:33:48 <Kepplar> do i remove it?
2006-08-21T12:33:56 <Kepplar> "use cache_dir as a base dir"
2006-08-21T12:33:58 <Kepplar> no shit :P
2006-08-21T12:34:01 <xorAxAx> oh, he did
2006-08-21T12:35:00 <Kepplar> oh that
2006-08-21T12:35:06 <Kepplar> i physically feel sick
2006-08-21T12:35:09 <Kepplar> cache_dir + '/pages/' + quoteWikinameFS(name)
2006-08-21T12:35:12 <Kepplar> thats filthy hack :/
2006-08-21T12:35:53 <birkenfeld> yes, he could use os.path.join
2006-08-21T12:35:57 <Kepplar> is there a way to work out if someones importing a module?
2006-08-21T12:36:12 <Kepplar> birkenfeld: i mean't more the fact its hardcoded =)
2006-08-21T12:37:59 <Kepplar> I would ideally like a way to say bog off to any imports from unauthorised modules
2006-08-21T12:47:30 <xorAxAx> no, thats not possible
2006-08-21T12:47:39 <xorAxAx> but at call time, you could check that
2006-08-21T12:47:48 <xorAxAx> you would just need a meta class that ensures that
2006-08-21T13:03:32 <Kepplar> whats pyxml and why do i need it?
2006-08-21T13:05:27 <xorAxAx> its needed because the xml code included in python is partly broken
2006-08-21T13:05:30 <xorAxAx> pyxml replaces it
2006-08-21T13:05:46 <richardb> http://pyxml.sourceforge.net/
2006-08-21T13:05:48 <xorAxAx> but moin runs also without it
2006-08-21T13:06:45 <Kepplar> o?
2006-08-21T13:06:52 <Kepplar> how do i disable it then?
2006-08-21T13:06:58 <xorAxAx> just dont install it
2006-08-21T13:07:09 <Kepplar> raise InternalError("You need to install PyXML to use the DocBook formatter.
2006-08-21T13:07:12 <Kepplar> ")
2006-08-21T13:07:14 <Kepplar> InternalError: You need to install PyXML to use the DocBook formatter.
2006-08-21T13:08:19 <xorAxAx> yep
2006-08-21T13:08:33 <xorAxAx> did you use the docbook formatter explicitly?
2006-08-21T13:10:48 <Kepplar> no
2006-08-21T13:11:00 <Kepplar> testSyntaxReference
2006-08-21T13:11:04 <Kepplar> which is a pretty critical test for me
2006-08-21T13:11:45 <xorAxAx> then change the test to import pyxml conditionally
2006-08-21T13:11:59 <xorAxAx> umm, to skip the docbook formatter if it raises an internalerror
2006-08-21T13:13:09 <Kepplar> kk
2006-08-21T13:31:24 <Kepplar> Traceback (most recent call last): File "/home/a/moin/storage/MoinMoin/_tests/test_PageEditor.py", line 10
2006-08-21T13:31:28 <Kepplar> 2, in testMailto self.assertEqual(self.expand(), u'[[MailTo(%s)]]' % self.email)
2006-08-21T13:31:31 <Kepplar> File "/home/a/moin/storage/MoinMoin/_tests/test_PageEditor.py", line 57
2006-08-21T13:31:33 <Kepplar> , in expand
2006-08-21T13:31:36 <Kepplar> return self.page._expand_variables(self.variable)
2006-08-21T13:31:39 <Kepplar> File "/home/a/moin/storage/MoinMoin/PageEditor.py", line 698, in _expan
2006-08-21T13:31:42 <Kepplar> d_variables
2006-08-21T13:31:44 <Kepplar> if self.request.dicts.has_dict(userDictPage):
2006-08-21T13:31:47 <Kepplar> File "/home/a/moin/storage/MoinMoin/request/__init__.py", line 239, in
2006-08-21T13:31:50 <Kepplar> getDicts
2006-08-21T13:31:52 <Kepplar> dicts.scandicts()
2006-08-21T13:31:55 <Kepplar> File "/home/a/moin/storage/MoinMoin/wikidicts.py", line 381, in scandic
2006-08-21T13:31:58 <Kepplar> ts
2006-08-21T13:32:00 <Kepplar> if Page.Page(request, pagename).mtime_usecs() >= self.pageupdate_timestamp:
2006-08-21T13:32:03 <Kepplar> File "/home/a/moin/storage/MoinMoin/Page.py", line 314, in mtime_usecs
2006-08-21T13:32:06 <Kepplar> for line in logfile.reverse():
2006-08-21T13:32:09 <Kepplar> File "/home/a/moin/storage/MoinMoin/logfile/__init__.py", line 102, in
2006-08-21T13:32:12 <Kepplar> reverse
2006-08-21T13:32:15 <Kepplar> self.to_end()
2006-08-21T13:32:17 <Kepplar> File "/home/a/moin/storage/MoinMoin/logfile/__init__.py", line 320, in
2006-08-21T13:32:20 <Kepplar> to_end
2006-08-21T13:32:22 <Kepplar> self._input.seek(0, 2) # to end of file
2006-08-21T13:32:25 <Kepplar> File "/home/a/moin/storage/MoinMoin/logfile/__init__.py", line 140, in
2006-08-21T13:32:28 <Kepplar> __getattr__
2006-08-21T13:32:31 <Kepplar> self._input = file(self.__filename, "rb",)
2006-08-21T13:32:33 <Kepplar> File "/home/a/moin/storage/MoinMoin/logfile/__init__.py", line 153, in
2006-08-21T13:32:36 <Kepplar> __getattr__
2006-08-21T13:32:39 <Kepplar> raise AttributeError(name)
2006-08-21T13:32:41 <Kepplar> AttributeError: _LogFile__filename
2006-08-21T13:32:44 <Kepplar> i have no clue how this works
2006-08-21T13:33:18 <xorAxAx> you commented out the code that sets the filename in the logfile object
2006-08-21T13:35:34 <Kepplar> doesnt look like it
2006-08-21T13:36:04 <Kepplar> thats a really wierd attribute name
2006-08-21T13:37:16 <xorAxAx> no, its a private attribute
2006-08-21T13:37:38 <Kepplar> but why the double undderscore between File and file
2006-08-21T13:37:39 <Kepplar> ?
2006-08-21T13:37:52 <Kepplar> it it were a dot it would make sense
2006-08-21T13:37:57 <xorAxAx> its obfuscated by python
2006-08-21T13:38:02 <xorAxAx> this is how private attributes work
2006-08-21T13:38:05 <Kepplar> yes
2006-08-21T13:38:11 <Kepplar> but not half way down the variable name
2006-08-21T13:38:20 <Kepplar> if its was double at the start it would make sense
2006-08-21T13:38:43 <Kepplar> _LogFile.__filename
2006-08-21T13:38:43 <xorAxAx> the reason is line 127-133 here: http://moin.pocoo.org:8080/moin-1.6-storage-aa?f=63537fa39d57;file=MoinMoin/logfile/editlog.py
2006-08-21T13:38:45 <Kepplar> would make sense
2006-08-21T13:39:00 <xorAxAx> Kepplar: please contact guido van rossum if you think that this would be better
2006-08-21T13:39:22 <Kepplar> its always been like that
2006-08-21T13:39:43 <Kepplar> your link is clipping
2006-08-21T13:40:06 <Kepplar> its always been the either _ or __ prefixing the variable to make it private
2006-08-21T13:40:16 <Kepplar> not letsnamethe__variable
2006-08-21T13:40:28 <xorAxAx> Kepplar: as i said above, this is done by python
2006-08-21T13:40:34 <xorAxAx> and explained by PEP8 IIRC
2006-08-21T13:41:48 <Kepplar> - __double_leading_underscore: when naming a class attribute, invokes name
2006-08-21T13:41:50 <Kepplar> mangling (inside class FooBar, __boo becomes _FooBar__boo; see below).
2006-08-21T13:41:53 <Kepplar> LEADING
2006-08-21T13:42:05 <xorAxAx> yes, __filename has 2 leading underscores
2006-08-21T13:42:29 <Kepplar> AttributeError: _LogFile__filename
2006-08-21T13:42:56 * xorAxAx reproduced reimars problem - it seems to be a bug in the merge code
2006-08-21T13:44:20 <Kepplar> oh
2006-08-21T13:44:22 <Kepplar> thats the RESULT
2006-08-21T14:10:31 <Kepplar> Grrr
2006-08-21T14:10:37 <Kepplar> sem in user apparently doesnt exist
2006-08-21T14:10:49 <Kepplar> created in constructor
2006-08-21T14:10:53 <Kepplar> and i know its there
2006-08-21T14:11:00 <Kepplar> then constructor calls another method
2006-08-21T14:11:02 <Kepplar> which can't see it!
2006-08-21T14:11:06 <Kepplar> attributeerror!
2006-08-21T14:30:44 <xorAxAx> fpletz: dont forget to document the new syntax on HelpOnSearching
2006-08-21T15:11:31 <mvirkkil> Reading the log i got nervous you had been crashing because of my code (docbook) but luckily that's not a bug but a feature ;)
2006-08-21T15:24:49 <fpletz`> moin
2006-08-21T15:25:07 <fpletz`> somehow my dyndns setup isn't working for my home machine.. so here I am
2006-08-21T15:26:49 <richardb> Hi
2006-08-21T15:27:49 <fpletz> I'm now working on the docs
2006-08-21T15:28:23 <richardb> Cool. I see you've fixed some of the problems with odd inputs to the advanced search UI.
2006-08-21T15:29:06 <fpletz> yes, and added some more stuff, like history searching
2006-08-21T15:29:23 <fpletz> unfortunately it's till pretty basic.. but, well :)
2006-08-21T15:29:33 <fpletz> *still
2006-08-21T15:29:46 <ThomasWaldmann> re
2006-08-21T15:29:51 <richardb> Have you fixed the "XXX" on the modified-since search box?
2006-08-21T15:30:09 <richardb> Should indicate what format of input is expected in the box.
2006-08-21T15:30:19 <richardb> Or does that not work yet?
2006-08-21T15:30:30 <fpletz> yes, this works now
2006-08-21T15:30:44 <fpletz> XXX meant that it didn't work yet :)
2006-08-21T15:30:50 <richardb> Ah.
2006-08-21T15:31:00 <fpletz> you can even input stuff like 'last 2 weeks" :)
2006-08-21T15:33:43 <xorAxAx> ThomasWaldmann: please fix moinmaster
2006-08-21T15:35:12 <fpletz> ThomasWaldmann: I'd like to second that :)
2006-08-21T15:35:41 <fpletz> works now :)
2006-08-21T15:36:03 <fpletz> uhm, I mean my dns stuff, moinmaster still broken
2006-08-21T15:37:47 <xorAxAx> hehe
2006-08-21T15:44:02 <ThomasWaldmann> that WikiDict page with was a bad idea
2006-08-21T15:44:21 <xorAxAx> ThomasWaldmann: oh, why is that?
2006-08-21T15:44:51 <xorAxAx> it only contains a redirect
2006-08-21T15:44:55 <ThomasWaldmann> i have to debug that
2006-08-21T15:45:17 <richardb> Back in 10 minutes. (need to move computer)
2006-08-21T15:50:32 <Kepplar> Hmm
2006-08-21T15:50:33 <Kepplar> *sigh*
2006-08-21T15:51:28 <Kepplar> ThomasWaldmann: any idea?
2006-08-21T15:52:54 <ThomasWaldmann> Kepplar: either you ask more concrete or you have to wait until i have found it in the backlog
2006-08-21T15:54:14 <xorAxAx> Kepplar: yeah, either show us the traceback or pose a clear question
2006-08-21T15:57:09 <ThomasWaldmann> Kepplar: are you still wondering about __filename in logfile or what's the problem?
2006-08-21T15:59:51 <xorAxAx> compile_huge_and_ugly = 0.421s getACL = 0.002s getPageList = 0.080s loadLanguage = 16.473s load_multi_cfg = 0.000s run = 17.035s send_page = 16.872s send_page_content = 0.101s total = 20.795s
2006-08-21T15:59:57 <xorAxAx> "loadLanguage = 16.473s"
2006-08-21T16:00:08 <xorAxAx> (thats MoinMaster)
2006-08-21T16:07:14 <Kepplar> back
2006-08-21T16:07:15 <Kepplar> sorry
2006-08-21T16:07:28 <Kepplar> nah it was the problem concerning the sem not existing when it was
2006-08-21T16:07:41 <Kepplar> it prodded it .. dont think i changed anything and it suddently worked
2006-08-21T16:07:48 <xorAxAx> prodded?
2006-08-21T16:08:33 <Kepplar> ThomasWaldmann: so what does this WikiDict stuff do, I have one of these tests requested for a Group using "" as the text, i think its leaking to storage as storage is trying to access a item with no name
2006-08-21T16:09:02 <Kepplar> xorAxAx: put print debugs in and played around with no obvious changes
2006-08-21T16:09:43 <xorAxAx> Kepplar: then the page doesnt exist of course
2006-08-21T16:09:51 <Kepplar> well i know that
2006-08-21T16:09:53 <xorAxAx> Kepplar: and you have to signal that
2006-08-21T16:09:58 <Kepplar> ah
2006-08-21T16:10:09 <Kepplar> that text field is suppose to be the Page name O-o
2006-08-21T16:10:13 <Kepplar> for a group?
2006-08-21T16:10:20 <xorAxAx> no idea
2006-08-21T16:15:40 <Kepplar> Group in WikiDict uses a dictionary when it should just use a list from what I can see....
2006-08-21T16:16:54 <birkenfeld> the deadline is in 3/4 h, IIRC?
2006-08-21T16:17:03 <Kepplar> probably
2006-08-21T16:18:27 <ThomasWaldmann> Kepplar: wikidicts should work with no problem if caching and raw page text access works
2006-08-21T16:18:54 <ThomasWaldmann> (except if you have strange WikiDict pages with #REDIRECT :)
2006-08-21T16:19:06 <Kepplar> i have a feeling the tests are trying to access a page which doesnt exist
2006-08-21T16:19:20 <Kepplar> and thus throws and exception
2006-08-21T16:19:23 <Kepplar> how to handle it i dont know
2006-08-21T16:19:23 <xorAxAx> ThomasWaldmann: i think the problem is that the dict was listed in a group page
2006-08-21T16:19:26 <Kepplar> what oes it expect?
2006-08-21T16:19:39 <Kepplar> d
2006-08-21T16:19:49 <xorAxAx> Kepplar: its ok to create pages that dont exist
2006-08-21T16:19:56 <xorAxAx> Kepplar: exists() returns False, though
2006-08-21T16:20:02 <Kepplar> indeed
2006-08-21T16:20:06 <ThomasWaldmann> a page with a first level itemize list with names
2006-08-21T16:20:16 <xorAxAx> (how should you create new pages if you are not allowed to create non-existing ones?)
2006-08-21T16:20:20 <Kepplar> eh?
2006-08-21T16:20:31 <xorAxAx> fpletz: have you enhanced HelpOnSearching already?
2006-08-21T16:20:42 <Kepplar> all i know is its trying to open a page ""
2006-08-21T16:20:54 <xorAxAx> thats perfectly legal
2006-08-21T16:21:20 <Kepplar> xorAxAx: i can't tell without docstrings telling me the return type :P
2006-08-21T16:21:30 <Kepplar> what does it return?
2006-08-21T16:21:35 <xorAxAx> Kepplar: a constructor returns an object of its type
2006-08-21T16:21:37 <Kepplar> what should get_raw() return for an invalid apge
2006-08-21T16:21:45 <Kepplar> page*
2006-08-21T16:21:53 <xorAxAx> get_raw doesnt exist
2006-08-21T16:21:55 <Kepplar> xorAxAx: im not talking about a constructor
2006-08-21T16:22:03 <fpletz> xorAxAx: yes
2006-08-21T16:22:09 <xorAxAx> cool
2006-08-21T16:22:11 <Kepplar> get_raw_body()
2006-08-21T16:22:45 <Kepplar> An attempted access for Page.get_raw_body() on a non-existant page should return what?
2006-08-21T16:22:57 <xorAxAx> that returns u"" for empty pages, you may want to raise an exeption in your backend, though
2006-08-21T16:23:12 <Kepplar> it already raises an exception
2006-08-21T16:23:18 <Kepplar> ok
2006-08-21T16:23:22 <Kepplar> I'll get Page to handle it
2006-08-21T16:24:13 <Kepplar> As I don't believe in the future it should be like that - u"" could be an existant blank page, external code check before calling
2006-08-21T16:24:18 <Kepplar> infact
2006-08-21T16:24:32 <xorAxAx> moin doesnt support empty pages
2006-08-21T16:24:39 <Kepplar> oh
2006-08-21T16:24:45 <Kepplar> yet =)
2006-08-21T16:26:02 <Kepplar> "" is invalid page anyway
2006-08-21T16:26:25 <Kepplar> I should get retrive_page on the sem to except
2006-08-21T16:28:12 <Kepplar> I think a regex for valid names will work
2006-08-21T16:28:26 <Kepplar> is it just \W+
2006-08-21T16:28:26 <Kepplar> ?
2006-08-21T16:28:33 <Kepplar> er
2006-08-21T16:28:42 <Kepplar> nah it supports \w
2006-08-21T16:28:48 <Kepplar> well some
2006-08-21T16:28:49 <Kepplar> hmm
2006-08-21T16:28:52 <Kepplar> what do you think?
2006-08-21T16:29:16 <xorAxAx> all names are valid
2006-08-21T16:29:28 <Kepplar> "" isnt
2006-08-21T16:29:31 <xorAxAx> some may yield a different name through normalisation, though
2006-08-21T16:29:55 <Kepplar> still
2006-08-21T16:29:57 <Kepplar> "" can't be
2006-08-21T16:34:00 <mvirkkil> ThomasWaldmann: I was planning on putting my documentation up at fedora's wiki, but come to think of it, somewhere more closely related to moin would probably be more appropriate.
2006-08-21T16:34:04 <mvirkkil> Suggestions?
2006-08-21T16:37:15 <xorAxAx> mvirkkil: see MoinMoin:EditingOnMoinMaster
2006-08-21T16:49:15 <dreimark> I do have some questions for the xapian search too. Is there a common place to add them.
2006-08-21T16:53:18 <xorAxAx> ThomasWaldmann: the problem is back
2006-08-21T16:53:20 <xorAxAx> ThomasWaldmann: http://moinmoin.wikiwikiweb.de/
2006-08-21T16:53:27 <xorAxAx> i didnt create a new page :)
2006-08-21T16:53:39 <xorAxAx> ThomasWaldmann: remove it from SystemPagesInEnglishGroup
2006-08-21T16:53:44 <xorAxAx> ThomasWaldmann: with vim
2006-08-21T16:53:50 <xorAxAx> ThomasWaldmann: do not prune WikiDict
2006-08-21T16:57:56 <xorAxAx> ThomasWaldmann: did you change something?
2006-08-21T16:58:09 <ThomasWaldmann> i renamed WikiDict to WikiDictXXX
2006-08-21T16:59:59 <xorAxAx> ok
2006-08-21T17:02:15 <xorAxAx> dreimark: see FeatureRequests/AdvancedXapianSearch
2006-08-21T17:12:58 <Kepplar> ah here we go the bugs back
2006-08-21T17:13:08 <Kepplar> Traceback (most recent call last): File "/home/a/moin/storage/MoinMoin/_tests/test_user.py", line 90, in t
2006-08-21T17:13:12 <Kepplar> estAsciiPassword theUser = user.User(self.request, name=name, password=password)
2006-08-21T17:13:14 <Kepplar> File "/home/a/moin/storage/MoinMoin/user.py", line 281, in __init__
2006-08-21T17:13:17 <Kepplar> self.id = getUserId(self._request, self.name)
2006-08-21T17:13:20 <Kepplar> File "/home/a/moin/storage/MoinMoin/user.py", line 77, in getUserId
2006-08-21T17:13:22 <Kepplar> name = User(request, id=userid).name
2006-08-21T17:13:25 <Kepplar> File "/home/a/moin/storage/MoinMoin/user.py", line 277, in __init__
2006-08-21T17:13:27 <Kepplar> self.load_from_id(check_pass)
2006-08-21T17:13:30 <Kepplar> File "/home/a/moin/storage/MoinMoin/user.py", line 366, in load_from_id
2006-08-21T17:13:33 <Kepplar> if not self.exists():
2006-08-21T17:13:35 <Kepplar> File "/home/a/moin/storage/MoinMoin/user.py", line 348, in exists
2006-08-21T17:13:38 <Kepplar> self.retrieve_user()
2006-08-21T17:13:40 <Kepplar> File "/home/a/moin/storage/MoinMoin/user.py", line 305, in retrieve_use
2006-08-21T17:13:43 <Kepplar> r
2006-08-21T17:13:46 <Kepplar> self._item = self._sem.retrieve_item(wikiutil.armor_storage_request(self.id,
2006-08-21T17:13:49 <Kepplar> "user"), "flatfile-data")
2006-08-21T17:13:51 <Kepplar> AttributeError: User instance has no attribute '_sem'
2006-08-21T17:13:54 <Kepplar> self._sem gets set in the construtor!
2006-08-21T17:13:56 <Kepplar> and i put a print statement in and the object is there!
2006-08-21T17:13:59 <Kepplar> and nothing assigns on it!
2006-08-21T17:20:11 <xorAxAx> Kepplar: obviously, the call chain goes up to the constructor
2006-08-21T17:20:19 <xorAxAx> Kepplar: so you set _sem too late
2006-08-21T17:25:10 <Kepplar> er
2006-08-21T17:25:15 <Kepplar> its calling exist() on an object
2006-08-21T17:25:19 <Kepplar> which is already made
2006-08-21T17:25:22 <Kepplar> constructor is first
2006-08-21T17:25:50 <xorAxAx> no, the constructor is calling exist on itself
2006-08-21T17:25:57 <Kepplar> ah
2006-08-21T17:25:59 <Kepplar> ooo i see
2006-08-21T17:26:01 <Kepplar> hmm
2006-08-21T17:26:03 <Kepplar> nicely spotted
2006-08-21T17:26:04 <Kepplar> ;)
2006-08-21T17:26:07 <Kepplar> thx
2006-08-21T17:49:46 <ThomasWaldmann> mvirkkil: your last commit is 29h ago. if you have something to commit, do it ASAP.
2006-08-21T17:50:08 <ThomasWaldmann> Kepplar: your last commit is 7h ago, same thing.
2006-08-21T17:50:26 <ThomasWaldmann> xorAxAx: your's 69 minutes :)
2006-08-21T17:50:45 <ThomasWaldmann> fpletz: your's 32 minutes :)
2006-08-21T17:50:49 <xorAxAx> yeah, i wont commit anything today anymore i think :)
2006-08-21T17:51:08 <xorAxAx> i need to get this multicall reactor working :)
2006-08-21T17:51:17 <ThomasWaldmann> so esp. kepplar and mvirkkil maybe want
2006-08-21T17:51:22 <fpletz> heh, me neither ;)
2006-08-21T17:53:27 <ThomasWaldmann> fpletz:
2006-08-21T17:53:27 <ThomasWaldmann> (16:53) < xorAxAx> i didnt create a new page :)
2006-08-21T17:53:36 <ThomasWaldmann> ignore
2006-08-21T17:53:47 <ThomasWaldmann> fpletz: Your search query "(test)" is invalid. Please refer to HelpOnSearch for more information.
2006-08-21T17:54:05 <ThomasWaldmann> i just entered "test" (without quotes) and hit enter
2006-08-21T17:54:09 <fpletz> uh
2006-08-21T17:54:12 <fpletz> a sec
2006-08-21T17:55:27 <fpletz> ThomasWaldmann: where did you enter it?
2006-08-21T17:55:34 <fpletz> works for me
2006-08-21T17:55:47 <ThomasWaldmann> on AdvancedSearch on xapian.www.de
2006-08-21T17:56:02 * fpletz smells py2.3
2006-08-21T17:56:16 <ThomasWaldmann> yes, it is py 2.3 :)
2006-08-21T17:56:23 <mvirkkil> ThomasWaldmann: Why?
2006-08-21T17:56:50 <fpletz> had problems with sort today.. the __doc__ on that is wrong :/
2006-08-21T17:56:57 <fpletz> but only in py2.3
2006-08-21T17:58:08 <fpletz> hmmm
2006-08-21T17:58:35 <fpletz> ah
2006-08-21T17:58:39 <fpletz> I know ;)
2006-08-21T17:58:47 <fpletz> please rebuild the index
2006-08-21T17:59:42 <birkenfeld> fpletz: entering ".*" in the second field still gives a traceback
2006-08-21T17:59:45 <xorAxAx> fpletz: the kwargs handling looked weird
2006-08-21T17:59:53 <xorAxAx> fpletz: sort accepts nearly no kwargs
2006-08-21T18:00:24 <fpletz> xorAxAx: in py2.3 it accepts no kwargs altough advertised in __doc__... if I had known that beforehand
2006-08-21T18:00:29 <fpletz> but that works now
2006-08-21T18:00:49 <xorAxAx> fpletz: it accepts cmpfunc there
2006-08-21T18:00:55 <fpletz> birkenfeld: mmh, I thought i fixed that too.. sec..
2006-08-21T18:00:57 <xorAxAx> fpletz: nothing else
2006-08-21T18:01:06 <birkenfeld> xorAxAx: it is no keyword argument
2006-08-21T18:01:09 <fpletz> xorAxAx: no, it doesn't even accept cmpfunc
2006-08-21T18:01:24 <birkenfeld> this is a general problem with functions implemented in C...
2006-08-21T18:01:26 <xorAxAx> birkenfeld: oh, that is confusing
2006-08-21T18:01:29 <birkenfeld> yep
2006-08-21T18:01:32 <xorAxAx> fpletz: ok, thats bad
2006-08-21T18:01:37 <xorAxAx> evil birkenfeld :)
2006-08-21T18:02:02 <xorAxAx> birkenfeld: not all advertise their signature like [].sort
2006-08-21T18:02:15 <fpletz> birkenfeld: you testet that on test.www.de, right? please test on xapian.www.de ;)
2006-08-21T18:02:31 <birkenfeld> fpletz: no, on xapian.www.de
2006-08-21T18:02:33 <fpletz> *tested
2006-08-21T18:02:37 <fpletz> mmh
2006-08-21T18:02:53 <birkenfeld> it works now
2006-08-21T18:03:12 <fpletz> ah, ok
2006-08-21T18:03:23 <fpletz> I was sure I fixed that :)
2006-08-21T18:04:26 <birkenfeld> hmmm...
2006-08-21T18:04:29 <birkenfeld> fpletz: http://xapian.wikiwikiweb.de/AdvancedFindPage?action=fullsearch&advancedsearch=1&and_terms=.*&or_terms=¬_terms=&categories=&mtime=&language=&mimetype=&includeunderlay=1
2006-08-21T18:05:17 <fpletz> yes, the index is outdated
2006-08-21T18:05:27 <fpletz> ThomasWaldmann: please rebuild :)
2006-08-21T18:06:07 <xorAxAx> fpletz: why isnt that possible using an action?
2006-08-21T18:06:26 <xorAxAx> "Please refer to HelpOnSearch for more information." it is helponsearching
2006-08-21T18:06:31 <fpletz> yes
2006-08-21T18:06:33 <fpletz> just fixed :)
2006-08-21T18:06:53 <xorAxAx> and "only system pages" is still very useless
2006-08-21T18:07:05 <xorAxAx> "no system pages" would make sense
2006-08-21T18:07:14 <fpletz> hmm
2006-08-21T18:07:26 <xorAxAx> from a usability PoV
2006-08-21T18:07:30 <fpletz> it doesn't work because the index is out of date
2006-08-21T18:07:41 <xorAxAx> how does it notice that?
2006-08-21T18:07:50 <birkenfeld> hm
2006-08-21T18:07:51 <xorAxAx> (and why is the error message incorrect? %-)
2006-08-21T18:08:09 <fpletz> I know because I changed some keys and terms :)
2006-08-21T18:08:11 <birkenfeld> "since last 2 weeks" is somehow wrong...
2006-08-21T18:10:06 <birkenfeld> http://xapian.wikiwikiweb.de/AdvancedFindPage?action=fullsearch&advancedsearch=1&and_terms=%28%29&or_terms=¬_terms=&categories=&mtime=&language=&mimetype=&includeunderlay=1
2006-08-21T18:12:40 <xorAxAx> fpletz: cant it say "invalid index" or do a fallback?
2006-08-21T18:13:02 <birkenfeld> still giving a traceback
2006-08-21T18:16:33 <fpletz> ok, pushed some fixes
2006-08-21T18:17:38 <fpletz> hmm
2006-08-21T18:19:00 <fpletz> xorAxAx: determining if the index is invalid would be possible but shouldn't happen in practice because one doesn't change things like sortkeys in minor versions
2006-08-21T18:19:18 <xorAxAx> fpletz: but in major versions. and then?
2006-08-21T18:19:19 <fpletz> but that should be added anway, yes
2006-08-21T18:19:26 <xorAxAx> fpletz: you need to store a serial number
2006-08-21T18:19:46 <fpletz> yup, that should do it
2006-08-21T18:20:44 <fpletz> birkenfeld: did you mean by "still" that you rebuilt the index or something? I can't do anything about that
2006-08-21T18:20:57 <fpletz> but the other issues should be fixed
2006-08-21T18:21:05 <xorAxAx> fpletz: you could add an action to do that
2006-08-21T18:21:18 <xorAxAx> that would require superuser permissions
2006-08-21T18:22:23 <fpletz> xorAxAx: mmh, right, that would be a lot more convenient..
2006-08-21T18:22:43 <ThomasWaldmann> fpletz: ran rebuild, no change
2006-08-21T18:22:58 <ThomasWaldmann> index build --files=files.lst --mode=rebuild
2006-08-21T18:23:28 <birkenfeld> why does it tell me that I searched for "(test)" ?
2006-08-21T18:25:09 <fpletz> birkenfeld: because the terms in the textfields get grouped with () and and'ed
2006-08-21T18:25:15 <fpletz> ThomasWaldmann: checking..
2006-08-21T18:26:00 <birkenfeld> ah
2006-08-21T18:27:14 <ThomasWaldmann> the language dropdown does not have "German" at 2nd place
2006-08-21T18:27:41 <ThomasWaldmann> (but english)
2006-08-21T18:27:51 <ThomasWaldmann> (but the wiki talks german to me)
2006-08-21T18:28:22 <ThomasWaldmann> (I am not logged in there)
2006-08-21T18:28:34 <fpletz> the 2nd place is either user.language or, if false, cfg.language_default
2006-08-21T18:28:38 <ThomasWaldmann> (I use too many brackets)
2006-08-21T18:29:27 <xorAxAx> fpletz: request.language or something like that
2006-08-21T18:29:47 <xorAxAx> request.lang
2006-08-21T18:29:53 <fpletz> okay, will use this :)
2006-08-21T18:30:00 <fpletz> thanks
2006-08-21T18:30:42 <ThomasWaldmann> Your search query "(or or and or or or not)" is invalid. Please refer to HelpOnSearching for more information.
2006-08-21T18:30:50 <ThomasWaldmann> (I am mean :)
2006-08-21T18:33:38 <Kepplar> i messed something up
2006-08-21T18:33:50 <Kepplar> how do i on my machine find out what files changed on last commit?
2006-08-21T18:34:33 <ThomasWaldmann> hg diff --help
2006-08-21T18:35:54 <xorAxAx> hg status
2006-08-21T18:36:09 <ThomasWaldmann> "on", not "since"
2006-08-21T18:37:49 <ThomasWaldmann> 50 #if self.locking:
2006-08-21T18:37:50 <ThomasWaldmann> 51 # self.lock_dir = os.path.join(self.arena_dir, '__lock__')
2006-08-21T18:37:50 <ThomasWaldmann> 52 # self.rlock = lock.ReadLock(self.lock_dir, 60.0)
2006-08-21T18:37:50 <ThomasWaldmann> 53 # self.wlock = lock.WriteLock(self.lock_dir, 60.0)
2006-08-21T18:38:00 <ThomasWaldmann> it is rather clear why it doesnt find wlock
2006-08-21T18:38:08 <Kepplar> i know that
2006-08-21T18:38:12 <Kepplar> fixed :P
2006-08-21T18:38:15 <Kepplar> well
2006-08-21T18:38:20 <Kepplar> back to where we were
2006-08-21T18:38:23 <Kepplar> sorta
2006-08-21T18:39:43 <Kepplar> ill push waht i have in a min brb loo
2006-08-21T18:42:43 <xorAxAx> birkenfeld: any experience in threaded programming?
2006-08-21T18:42:49 <xorAxAx> birkenfeld: its damn hard to get it right :)
2006-08-21T18:47:14 <xorAxAx> hmm, at least domain could be implemented for the normal search as well
2006-08-21T18:48:26 <xorAxAx> fpletz: did you optimise the paging/sorting?
2006-08-21T18:51:35 <xorAxAx> fpletz: the page counters need more hspace
2006-08-21T18:51:45 <xorAxAx> 12345678910Next
2006-08-21T18:51:47 <xorAxAx> %-)
2006-08-21T18:52:00 <xorAxAx> i.e. some left and right padding
2006-08-21T19:00:56 <fpletz> ThomasWaldmann: don't index any files for now, that seems to be broken (never tested that)
2006-08-21T19:01:22 <fpletz> xorAxAx: the other searches too.. wouldn't be too hard
2006-08-21T19:01:35 <fpletz> xorAxAx: paging/sorting is partly handled by xapian
2006-08-21T19:01:46 <fpletz> xorAxAx: uhm.. links/lynx? :)
2006-08-21T19:19:50 <ThomasWaldmann> fpletz: i removed files.lst and it still says the same
2006-08-21T19:19:57 <ThomasWaldmann> http://xapian.wikiwikiweb.de/AdvancedFindPage?action=fullsearch&advancedsearch=1&and_terms=test&or_terms=¬_terms=&categories=&mtime=&language=&mimetype=&includeunderlay=1
2006-08-21T19:20:21 <ThomasWaldmann> (and deleted whole xapian index and rebuilt it)
2006-08-21T19:20:56 <ThomasWaldmann> btw, why didn't you run those test yourself and earlier?
2006-08-21T19:20:59 <fpletz> but why does it work for me locally.. let's test something else..
2006-08-21T19:21:19 <fpletz> worked for me :)
2006-08-21T19:27:56 <ThomasWaldmann> xorAxAx: i guess i have found the WikiDict bug
2006-08-21T19:28:04 <ThomasWaldmann> you may try to crash it again
2006-08-21T19:29:16 <ThomasWaldmann> while trying to expand the hierarchical groups, it tried to expand into WikiDict and that failed because of missing members() method
2006-08-21T19:29:34 <ThomasWaldmann> but it shouldnt try to expand into dicts anyway
2006-08-21T19:38:02 <fpletz> ThomasWaldmann: works
2006-08-21T19:39:38 <ThomasWaldmann> http://hg.thinkmo.de/moin/1.6-xapian-fpletz?cs=ad6bfa51d16c does someone see the diff for longer than 50ms?
2006-08-21T19:40:39 <fpletz> mmh, yes
2006-08-21T19:41:05 <ThomasWaldmann> with firefox. it shows up for a short time and then vanishes... %-)
2006-08-21T19:41:29 <ThomasWaldmann> http://xapian.wikiwikiweb.de/AdvancedFindPage?action=fullsearch&advancedsearch=1&and_terms=&or_terms=test¬_terms=&categories=&mtime=last+day&language=&mimetype=&includeunderlay=1
2006-08-21T19:41:59 <fpletz> not in my code :)
2006-08-21T19:43:21 <fpletz> mmh, why is the revision not set..
2006-08-21T19:52:30 <ThomasWaldmann> fpletz: the 1 2 3 links are incorrect after 2nd click, i guess you need some relative=False somewhere
2006-08-21T19:52:55 <ThomasWaldmann> (this is because of a recent change of me in moin-main)
2006-08-21T19:54:53 <ThomasWaldmann> http://xapian.wikiwikiweb.de/TestingOldRev?action=fullsearch&context=180&value=old+stuff&fullsearch=Text#preview
2006-08-21T19:55:17 <ThomasWaldmann> if you click on first link, it doesn't show correct (old) rev
2006-08-21T19:57:04 <ThomasWaldmann> and it is not rev 2 it found the old stuff in, but rev 1
2006-08-21T19:57:15 <ThomasWaldmann> you can add a &rev=1 to url
2006-08-21T19:59:39 <ThomasWaldmann> btw, if the small search box at top right searches all page revs by default, maybe there should also be the same default behaviour in AdvancedSearch macro
2006-08-21T20:00:17 <ThomasWaldmann> (--> check last item OR only search current stuff in simple search)
2006-08-21T20:01:56 <ThomasWaldmann> Hi Paul,
2006-08-21T20:01:56 <ThomasWaldmann> All students should have completed work on their projects, meaning
2006-08-21T20:01:56 <ThomasWaldmann> their code should be available to their mentors for a final review and
2006-08-21T20:01:56 <ThomasWaldmann> mentors should evaluate their students' progress for the final
2006-08-21T20:01:56 <ThomasWaldmann> evaluation based on work done before this deadline.
2006-08-21T20:01:59 <ThomasWaldmann> HTH,
2006-08-21T20:02:01 <ThomasWaldmann> LH
2006-08-21T20:02:24 <ThomasWaldmann> (he asked about Aug 21, 0800 pacific time)
2006-08-21T20:03:23 <ThomasWaldmann> if we are nitpicking that, we can discuss that "should" <g>
2006-08-21T20:04:14 <ThomasWaldmann> bbl
2006-08-21T20:15:11 <birkenfeld> xorAxAx: I know, which is why I never did much with threads ;)
2006-08-21T21:05:51 <birkenfeld> I just found 6 obvious bugs in flatfile.py and 3 in storage.py ...
2006-08-21T21:11:26 <ThomasWaldmann> some stuff was even found by vim highlighting :)
2006-08-21T21:21:23 <mvirkkil> The worst bugreport I've ever submitted. Sorry about that guys: http://moinmoin.wikiwikiweb.de/MoinMoinBugs/CodeAreaNotAborted
2006-08-21T21:24:04 <Kepplar> obvious bugs?
2006-08-21T21:24:11 <Kepplar> btw
2006-08-21T21:24:13 <Kepplar> i pushed
2006-08-21T21:25:34 <Kepplar> birkenfeld?
2006-08-21T21:25:39 <Kepplar> brb
2006-08-21T21:25:45 <Kepplar> (wireless is choppy)
2006-08-21T21:25:46 <ThomasWaldmann> mvirkkil: the preparsing is rather simple and filters out ##comment
2006-08-21T21:26:48 <ThomasWaldmann> so the wiki parser never gets the line that begins with ##
2006-08-21T21:27:13 <mvirkkil> ThomasWaldmann: Yeah, I had a case which was a lot more complex, but I reduced it to that, and it seemed pretty clear what the problem was.
2006-08-21T21:27:30 <mvirkkil> Unfortunately it also a pretty evident bug.
2006-08-21T21:27:50 <ThomasWaldmann> it's by design
2006-08-21T21:28:18 <mvirkkil> Uuh.. But if it wouldn't get to the formatter, you wuoldn't see it on that page.
2006-08-21T21:28:18 <ThomasWaldmann> every line that begins with ## is a comment and not seen by the real parser
2006-08-21T21:28:32 <mvirkkil> But that one is seen, and feeded to the formatter, it seems.
2006-08-21T21:28:52 <ThomasWaldmann> hmmm, right. strange.
2006-08-21T21:29:20 <ThomasWaldmann> it happens in 1.6 also?
2006-08-21T21:29:24 <mvirkkil> What seems to happen is that it is seen, but not parsed. And it should be parsed.
2006-08-21T21:29:28 <mvirkkil> ThomasWaldmann: yes
2006-08-21T21:29:38 <mvirkkil> Since it can contain a }}}
2006-08-21T21:30:07 <mvirkkil> and there isn't supposed to be any markup that isn't allowed inside a }}}, except }}}
2006-08-21T21:30:19 <mvirkkil> inside a {{{ }}}
2006-08-21T21:41:27 <Kepplar> wheres the unicode to x translation generic function located?
2006-08-21T21:41:51 <xorAxAx> fpletz: nope, opera
2006-08-21T21:42:01 <xorAxAx> fpletz: the numbers are directly near each other
2006-08-21T21:42:07 <xorAxAx> it looks horrible
2006-08-21T21:42:14 <xorAxAx> i put it onto my todo list
2006-08-21T21:42:46 <xorAxAx> Kepplar: for what do you need it?
2006-08-21T21:43:28 <Kepplar> translation unicode data to something saveable
2006-08-21T21:43:34 <mvirkkil> unistring = "öötöntä menoa"
2006-08-21T21:44:00 <Kepplar> ?
2006-08-21T21:44:08 <mvirkkil> latinstring = unistring.encode('iso-8859-1')
2006-08-21T21:44:30 <mvirkkil> Kepplar: Why wouldn't unicode data be saveable?
2006-08-21T21:44:33 <ThomasWaldmann> mvirkkil: there are 2 places handling }}} and it seems inconsistent
2006-08-21T21:44:40 <Kepplar> mvirkkil: disk doesnt support unicode
2006-08-21T21:45:05 <mvirkkil> Kepplar: well, you can always use the encoding escape-string iirc
2006-08-21T21:45:24 <Kepplar> oh and is the edit-log global again?
2006-08-21T21:45:30 <Kepplar> Hmm
2006-08-21T21:45:46 <ThomasWaldmann> Kepplar: i guess it was in Page._write_file() # if you mean page content
2006-08-21T21:46:51 <mvirkkil> Kepplar: googling for python encoding conversion gives about a million hits.
2006-08-21T21:47:11 <Kepplar> mvirkkil: im thinking in relation to the project
2006-08-21T21:47:13 <xorAxAx> Kepplar: the text class has to have a function "convert2raw" r something like that
2006-08-21T21:47:16 <Kepplar> we should have one way of encoding
2006-08-21T21:47:23 <xorAxAx> Kepplar: as i said, it calls a method on self that returns the encoding
2006-08-21T21:47:35 <xorAxAx> Kepplar: then you feed the encoding enc like this:
2006-08-21T21:47:43 <xorAxAx> return data.encode(enc)
2006-08-21T21:48:24 <Kepplar> hold on
2006-08-21T21:48:28 <Kepplar> let me get this straight
2006-08-21T21:48:32 <Kepplar> the quotewikifs stuff
2006-08-21T21:48:34 <Kepplar> is different?
2006-08-21T21:48:40 <Kepplar> as not all char are support as fs names?
2006-08-21T21:48:48 <Kepplar> (This isn't explicatly documented)
2006-08-21T21:49:00 <xorAxAx> quotewikifs is about page names
2006-08-21T21:49:04 <xorAxAx> what are you talking about?
2006-08-21T21:49:11 <Kepplar> raw data
2006-08-21T21:49:12 <xorAxAx> no, the fs nearly supports nothing
2006-08-21T21:49:17 <xorAxAx> so not the pagename?
2006-08-21T21:49:21 <Kepplar> indeed
2006-08-21T21:49:30 <Kepplar> i was just thinking of running down the same function
2006-08-21T21:49:36 <xorAxAx> yeah, then call it like i desciribed above
2006-08-21T21:49:41 <Kepplar> ok
2006-08-21T21:56:34 <Kepplar> hmm
2006-08-21T21:58:25 <Kepplar> it appears its automatically trying to convert unicode to asciii
2006-08-21T21:58:27 <Kepplar> and failing
2006-08-21T21:58:32 <Kepplar> return self.access_storage(str_location, (lambda fp_revision_file: fp_revision_file.write(str_data)), "a")
2006-08-21T21:58:35 <Kepplar> UnicodeEncodeError: 'ascii' codec can't encode character u'\xfc' in position 307: ordinal not in range(128)
2006-08-21T22:03:25 <xorAxAx> Kepplar: yep!
2006-08-21T22:03:34 <xorAxAx> Kepplar: so where is the call i talked about?
2006-08-21T22:04:55 <Kepplar> well its its doing it automatically theres no difference?
2006-08-21T22:05:02 <Kepplar> (i know its bad practice, i want to get it working first)
2006-08-21T22:05:05 <Kepplar> its still out of range
2006-08-21T22:05:55 <xorAxAx> no, as you can see, it doesnt work if you dont start thinking
2006-08-21T22:06:05 <xorAxAx> i.e. it only works for ascii
2006-08-21T22:06:08 <xorAxAx> which is useles
2006-08-21T22:06:08 <xorAxAx> s
2006-08-21T22:06:18 <birkenfeld> 7 bit should be enough for everyone!
2006-08-21T22:06:20 <xorAxAx> assuming that you are a laywer and want to print copyright symbols
2006-08-21T22:07:01 <Kepplar> xorAxAx: ive lost you complete
2006-08-21T22:07:02 <Kepplar> er
2006-08-21T22:07:03 <Kepplar> completly
2006-08-21T22:07:05 <Kepplar> (lag)
2006-08-21T22:07:53 <xorAxAx> Kepplar: you have to do the things i described above, you can NOT defer this
2006-08-21T22:08:14 <xorAxAx> obviously ...
2006-08-21T22:08:18 <Kepplar> the raw part?
2006-08-21T22:08:22 <Kepplar> (your making little sense)
2006-08-21T22:08:40 <birkenfeld> lol
2006-08-21T22:10:48 <xorAxAx> Kepplar: the convert2raw part, yes
2006-08-21T22:11:04 <xorAxAx> and dont use that method name
2006-08-21T22:11:07 <xorAxAx> it was just an example
2006-08-21T22:11:34 <xorAxAx> maybe call it convert37331unicode2raw
2006-08-21T22:12:57 <Kepplar> 20:47 < xorAxAx> Kepplar: the text class has to have a function "convert2raw" r something like that
2006-08-21T22:13:00 <Kepplar> 20:47 < Kepplar> we should have one way of encoding
2006-08-21T22:13:03 <Kepplar> 20:47 < xorAxAx> Kepplar: as i said, it calls a method on self that returns the encoding
2006-08-21T22:13:06 <Kepplar> 20:47 < xorAxAx> Kepplar: then you feed the encoding enc like this:
2006-08-21T22:13:08 <Kepplar> 20:47 < xorAxAx> return data.encode(enc)
2006-08-21T22:13:38 <xorAxAx> yes, back then, it was already the second time i tried to describe the solution to this problem
2006-08-21T22:13:56 <Kepplar> I'm guessing i'd need to convert it manually to raw before i return it?
2006-08-21T22:15:17 <Kepplar> ive never touched string encoding prior to this
2006-08-21T22:15:18 <Kepplar> btw
2006-08-21T22:15:33 <xorAxAx> yes, files just accept bytes
2006-08-21T22:15:48 <birkenfeld> logically
2006-08-21T22:16:10 <Kepplar> this isnt making alot of sense
2006-08-21T22:16:34 <Kepplar> its the equivilent of reading every 5 words of the bible
2006-08-21T22:16:58 <birkenfeld> even that makes a lot of sense to some people
2006-08-21T22:17:13 <Kepplar> you've done encoding though?
2006-08-21T22:17:40 <Kepplar> GOing on preicsely what xorAxAx has said, hes saying to create a function that returns an encoded version
2006-08-21T22:17:45 <birkenfeld> encoding? it's as simple as .encode() and .decode()
2006-08-21T22:17:50 <Kepplar> i know
2006-08-21T22:17:55 <Kepplar> but
2006-08-21T22:18:04 <Kepplar> its already doing that and failing
2006-08-21T22:18:04 <xorAxAx> Kepplar: easy:
2006-08-21T22:18:10 <birkenfeld> only thing to decide is which charset to use
2006-08-21T22:18:11 <xorAxAx> Kepplar: 'First earth without darkness of Spirit over waters. "Let there God was the 5 Day, called evening one said, firmament the separate waters." the waters firmament were it God And there day. "Let heavens one dry was'
2006-08-21T22:18:15 <xorAxAx> Kepplar: does it help you?
2006-08-21T22:18:26 <birkenfeld> oops
2006-08-21T22:18:30 <Kepplar> xorAxAx: lol
2006-08-21T22:18:35 <xorAxAx> birkenfeld: 'First earth without darkness of Spirit over waters. "Let there God was the 5 Day, called evening one said, firmament the separate waters." the waters firmament were it God And there day. "Let heavens one dry was'
2006-08-21T22:18:57 <birkenfeld> wow
2006-08-21T22:19:02 * birkenfeld is enlightened
2006-08-21T22:19:15 <xorAxAx> 22:16:34 < Kepplar> its the equivilent of reading every 5 words of the bible
2006-08-21T22:19:30 <birkenfeld> so where are the encode calls?
2006-08-21T22:19:41 <xorAxAx> Kepplar: the problem is that your storage backend doesnt know the item encoding
2006-08-21T22:20:20 <birkenfeld> just using utf-21 should be enough, shouldn't it?
2006-08-21T22:20:32 <Kepplar> unicode string is currently atempting to access file.. it auto converts to ascii and fails. xorAxAx says create a function to simply return the encoding before hand... I'm assuming some other manual raw conversion is needed because else its precisly the same.
2006-08-21T22:21:04 <birkenfeld> I guess I'll have to open another bottle of wine today.
2006-08-21T22:21:16 <xorAxAx> Kepplar: no, its not
2006-08-21T22:21:21 <Kepplar> xorAxAx: well that depends on the particular engine
2006-08-21T22:21:22 <xorAxAx> Kepplar: that function wont return "ascii" of course
2006-08-21T22:21:40 <xorAxAx> Kepplar: the engine should just digest raw data
2006-08-21T22:21:49 <xorAxAx> and that means str, not unicode
2006-08-21T22:22:02 <Kepplar> ah
2006-08-21T22:22:06 <Kepplar> now that makes sense :)
2006-08-21T22:22:53 <Kepplar> We have loads of wine :)
2006-08-21T22:23:02 <Kepplar> I like my port instead though :)
2006-08-21T22:23:12 <birkenfeld> we?
2006-08-21T22:23:23 <birkenfeld> you and your friend?
2006-08-21T22:23:30 <Kepplar> family house
2006-08-21T22:23:34 <xorAxAx> just for reference, here is genesis again, this time in the opposite order (maybe you can get some coding hints from there, Kepplar):
2006-08-21T22:23:38 <xorAxAx> '10 appear." and gathered waters And morning, was the so. the the were and And waters and the there 6 there And the called from and that light. be And face Goddeep; upon and The made'
2006-08-21T22:23:50 <birkenfeld> there's no wine in it
2006-08-21T22:27:17 <Kepplar> in the bottle?
2006-08-21T22:27:58 <birkenfeld> in genesis
2006-08-21T22:28:22 <xorAxAx> ok, let me check the newer parts of the book
2006-08-21T22:28:32 <Kepplar> xorAxAx: you christian?
2006-08-21T22:28:36 <xorAxAx> i love the kepplar-moses operator: [::5]
2006-08-21T22:28:52 <xorAxAx> Kepplar: does it matter?
2006-08-21T22:28:56 <birkenfeld> I thought he's Alexander
2006-08-21T22:28:56 <Kepplar> nop
2006-08-21T22:29:19 <Kepplar> oh i was just wondering
2006-08-21T22:29:21 <xorAxAx> Kepplar: yes, i am
2006-08-21T22:29:48 <Kepplar> cool
2006-08-21T22:30:11 <birkenfeld> Kepplar: you're Anglican?
2006-08-21T22:30:38 <Kepplar> officially athiest but Im not stupid enough to close my mind that much :)
2006-08-21T22:31:12 <Kepplar> I have an interest in religon and theology
2006-08-21T22:31:40 <birkenfeld> well, as long as you don't plan attacks on airplanes that's fine ;)
2006-08-21T22:31:44 <xorAxAx> currently, computer science would matter more
2006-08-21T22:32:04 <Kepplar> heh
2006-08-21T22:32:53 <Kepplar> Perhaps computer science has more holy wars then the rest of the world combined
2006-08-21T22:33:59 <xorAxAx> but the evaluation of your mentor wont depend on your choice between vim and emacs
2006-08-21T22:34:12 <Kepplar> choice?
2006-08-21T22:34:28 <Kepplar> who in their right mind will choose emacs :P
2006-08-21T22:34:57 <birkenfeld> depending on what your interpretation of "right mind" is, I might feel offended by now
2006-08-21T22:35:07 <birkenfeld> ;)
2006-08-21T22:35:15 <Kepplar> :o you use emacs?
2006-08-21T22:35:22 <birkenfeld> indeed
2006-08-21T22:35:29 <Kepplar> i feel sorry for you
2006-08-21T22:35:33 <Kepplar> ;D
2006-08-21T22:35:37 <Kepplar> jk
2006-08-21T22:40:14 <birkenfeld> Kepplar: what does list.append return?
2006-08-21T22:40:37 <Kepplar> of what?
2006-08-21T22:40:48 <Kepplar> oh
2006-08-21T22:40:55 <Kepplar> not sure if it does
2006-08-21T22:41:12 <Kepplar> or
2006-08-21T22:41:21 <Kepplar> it returns the list concatinated
2006-08-21T22:41:32 <xorAxAx> Kepplar: spawn a console and try it
2006-08-21T22:41:35 <xorAxAx> a python console
2006-08-21T22:42:06 <Kepplar> nowt
2006-08-21T22:42:21 <Kepplar> file/line number ;)?
2006-08-21T22:42:46 <birkenfeld> flatfile.py:376
2006-08-21T22:42:52 <Kepplar> cheers :)
2006-08-21T22:43:54 <Kepplar> cool
2006-08-21T22:43:55 <Kepplar> fixed
2006-08-21T22:44:13 <Kepplar> (not that it made any difference - function worked =) )
2006-08-21T22:44:57 <birkenfeld> perhaps because that particular branch was never reached
2006-08-21T22:45:30 <Kepplar> must of done
2006-08-21T22:45:35 <Kepplar> as acl was processing fine
2006-08-21T22:45:45 <Kepplar> which is reason that area exists
2006-08-21T22:45:54 <Kepplar> hmm
2006-08-21T22:45:58 <Kepplar> that really needs documenting
2006-08-21T22:46:18 <birkenfeld> and while you're at it, you want to look up the meaning of the second argument to os.mkdir()
2006-08-21T22:46:20 <Kepplar> and im not happy with parse/serialise data functionnames
2006-08-21T22:46:58 <Kepplar> Um
2006-08-21T22:47:01 <Kepplar> thats to create parents
2006-08-21T22:49:48 <birkenfeld> who said that?
2006-08-21T22:50:13 <xorAxAx> birkenfeld: his friend did, i think
2006-08-21T22:50:18 <Kepplar> ...
2006-08-21T22:50:20 <Kepplar> most references?
2006-08-21T22:50:31 <birkenfeld> most?
2006-08-21T22:50:39 <birkenfeld> there's only one reference, and that's the library manual
2006-08-21T22:51:18 <Kepplar> i actually have an oreileys book here :P
2006-08-21T22:51:54 <xorAxAx> oh, the scientific approach
2006-08-21T22:51:58 <Kepplar> lol
2006-08-21T22:52:03 <Kepplar> whats the problem?
2006-08-21T22:52:09 <birkenfeld> I thought you were a computer scientist, not a physicist
2006-08-21T22:52:46 <Kepplar> tbh I'm more of a software architecture thoerist
2006-08-21T22:53:16 <Kepplar> some follow larry, some follow guido
2006-08-21T22:53:18 <Kepplar> i follow gamma :P
2006-08-21T22:53:26 <xorAxAx> ah, gamma, who is that?
2006-08-21T22:53:41 <Kepplar> Design Patterns?
2006-08-21T22:54:14 <xorAxAx> hmm, i have never needed a so called design pattern in python
2006-08-21T22:54:31 <Kepplar> I wouldn't be so sure :)
2006-08-21T22:54:49 <Kepplar> A lot of them are common sense
2006-08-21T22:54:56 <birkenfeld> whatever
2006-08-21T22:54:57 <birkenfeld> http://docs.python.org/lib/os-file-dir.html#l2h-1619
2006-08-21T22:55:00 <xorAxAx> yeah, like the function call design pattern
2006-08-21T22:55:07 <xorAxAx> or the name binding design pattern
2006-08-21T22:55:14 <xorAxAx> then we have the loop design pattern
2006-08-21T22:55:20 <xorAxAx> or the closure design pattern
2006-08-21T22:55:23 <Kepplar> lol
2006-08-21T22:56:44 <Kepplar> O_o
2006-08-21T22:56:44 <Kepplar> eh?
2006-08-21T22:57:04 <birkenfeld> what?
2006-08-21T22:57:57 <Kepplar> oh dont tell me it doesnt exist..
2006-08-21T22:58:58 <xorAxAx> it does
2006-08-21T22:59:08 <xorAxAx> in MoinMoin.support .util or shutil
2006-08-21T22:59:35 <Kepplar> where did i get that notation from then O_o
2006-08-21T23:04:00 <birkenfeld> what Python version does Moin require?
2006-08-21T23:04:29 <Kepplar> 2.3
2006-08-21T23:04:48 <birkenfeld> hm, then there's quite a lot of legacy stuff that can be removed
2006-08-21T23:05:01 <Kepplar> what did you have in mind?
2006-08-21T23:05:22 <birkenfeld> most of util.filesys
2006-08-21T23:05:29 <Kepplar> ah
2006-08-21T23:05:33 <birkenfeld> and so on
2006-08-21T23:06:57 <xorAxAx> birkenfeld: we happily accept patches :)
2006-08-21T23:07:44 <Kepplar> right guys im heading to bed, ive had less than 12 hours sleep since friday
2006-08-21T23:08:21 <birkenfeld> xorAxAx: let's see
2006-08-21T23:09:57 <xorAxAx> Kepplar: thats a very good idea
2006-08-21T23:11:05 <Kepplar> (sorry if ive come across tetchy in the past few days)
2006-08-21T23:11:06 <Kepplar> nn
2006-08-21T23:11:12 <birkenfeld> hehe
2006-08-21T23:12:04 <xorAxAx> Kepplar: a good idea because it generally increases code quality
2006-08-21T23:25:23 <birkenfeld> data = self.access_storage(str_location, (lambda fp_revision_file: fp_revision_file.read().encode("utf8")), "r")
2006-08-21T23:25:27 <birkenfeld> doesn't look right MoinMoin: MoinMoinChat/Logs/moin-dev/2006-08-21 (last edited 2021-02-03 10:44:37 by Bind)
